According to our results, the rental prices in Sydney have increased over time regardless of where students live. We recommend that the university provide more affordable housing options to make education more accessible and affordable for students.
Overview
This data was sourced from a survey containing 25 variables and was completed by 2103 students in DATA1901 and DATA1001 (99% of the cohort)
Our research focused on the variables: rent per week in AUD, Length of commute to campus (minutes), Cohort. We classified our variables quantitative continuous, quantitative continuous and qualitative discrete.
library(tidyverse) # for box and linear model
library(plotly) # for pie chart
library(RColorBrewer) # for recolouring pie chart
library(ggthemes) # for ggplot theme
surveydata = read.csv("data1001_survey_data_2025_S1.csv")
surveydata = filter(surveydata, consent == "I consent to take part in the study")
surveydata = surveydata[surveydata$rent <= 2000, ]
surveydata = surveydata[surveydata$rent != 0, ]
surveydata = surveydata[surveydata$commute <= 180, ]Limitations
Limitations of this data is that it is only a representation of this cohort and not a population so it will not accurately represent the fluctuating rent prices and commute times. There is a significant difference in cohort sizes for each semester, which can impact our data due to the different sample sizes.
cohort_summary <- surveydata %>%
count(cohort)
plot_ly(cohort_summary,
labels = ~cohort,
values = ~n,
type = 'pie',
textinfo = 'percent',
textposition = 'auto',
marker = list(colors = brewer.pal(length(unique(surveydata$cohort)), "Oranges"))) %>%
layout(title = 'Distribution of Semesters',
showlegend = TRUE)Assumptions
It was assumed that people were honest and reasonable in their responses, and responses where responders did not consent to take part in the survey were excluded. Students responded with the accurate amount of rent they paid in AUD$ per week, and also an appropriate approximation of the time it takes to commute to campus in minutes. “$0” entries were assumed to be students who lived with parents/family, and were excluded from the data as they do not contribute to data about rent prices. Commute time was capped to 180 minutes (3 hours), as 3 hours is a generous upper limit for reasonable commutes.
How have rent prices changed between Semester 2 of last year (2024) and Semester 1 this year (2025) ?
ggplot(surveydata, aes(x = rent, y = cohort)) +
geom_boxplot() +
labs(x = "Rent",
y = "Semester") +
theme_solarized() +
scale_fill_solarized()
ggplot(surveydata, aes(x = commute, y = rent)) +
geom_point() +
geom_smooth(method = "lm", se = FALSE) +
labs(x = "Commute",
y = "Rent") +
theme_solarized() +
scale_fill_solarized()
model = lm(rent ~ commute, data = surveydata)
# Create residual plot
# Create residual plot
ggplot(model, aes(x = .fitted, y = .resid)) +
geom_point() +
geom_hline(yintercept = 0, linetype = "dashed", colour = "red") +
labs(x = "Fitted Value",
y = "Residual") +
theme_solarized() +
scale_fill_solarized()
[1]Welch, I. (2025, January 28). House prices to rise by 3.3%, units by 4.6% in 2025. KPMG. https://kpmg.com/au/en/home/media/press-releases/2025/01/house-and-unit-prices-to-rise-in-2025.html
The Acknowledgment section includes a list of group meetings (date and time and attendance), the contribution of each group member, and all resources used (eg url of stack overflow, url of Ed post, date and details of drop-in session with tutor, record of ChatGPT session with prompt)